A simple method of assessing premenstrual syndrome in large prospective studies.

OBJECTIVE: To evaluate a simple method of assessing incident premenstrual syndrome (PMS) in large prospective studies. STUDY
DESIGN: Participants included 135 Nurses' Health Study II members who first reported PMS by questionnaire in 2001 and 371 who never reported PMS (1989-2001). All completed a retrospective menstrual symptom questionnaire in 2002. We used responses to classify cases and noncases into diagnostic groups for comparison. Logistic regression was used to assess the effect of using various case and noncase classifications on risk estimates in analyses of age, calcium intake and PMS.
RESULTS: The menstrual symptom occurrence, timing and severity in women meeting established PMS criteria as assessed by retrospective questionnaire were essentially identical to those who also reported prospective symptom charting (our "gold standard"). Relative risks calculated using these 2 case groups were similar, while results using less restrictive case and noncase definitions were substantially attenuated.
CONCLUSION: Prospective reporting of initial PMS diagnosis followed by a short retrospective symptom questionnaire to confirm symptom experience can identify cases in large prospective studies and appears sensitive enough to identify risk factors for incident PMS.